Rizzolatti, G., & Destro, M. F. (2008). Mirror neurons. Scholarpedia, 3(1), 2055.
[Note that the main reading has been updated. Bonini et al (2022) was too hard to understand, Rizzolattti & Destro (2008) has been substituted for it]
What might be the functional role of the mirror neuron system? A series of hypotheses such as imitation, action understanding, intention understanding, and empathy have been put forward to explain the functional role of the mirror neurons. In addition to these, it has also been suggested that the mirror neuron system represents the basic neural mechanism from which language evolved.
Bonini, L., Rotunno, C., Arcuri, E., & Gallese, V. (2022). Mirror neurons 30 years later: implications and applications. Trends in Cognitive Sciences.
Optional:
Bandera, J. P., Marfil, R., Molina-Tanco, L., Rodriguez, J. A., Bandera, A., & Sandoval, F. (2007). Robot learning by active imitation. INTECH Open Access Publisher.
Cook, R., Bird, G., Catmur, C., Press, C., & Heyes, C. (2014). Mirror neurons: from origin to function. Behavioral and Brain Sciences, 37(02), 177-192.
Bonini, L., Rotunno, C., Arcuri, E., & Gallese, V. (2022). Mirror neurons 30 years later: implications and applications. Trends in Cognitive Sciences.
Please post early in the week, not just before the next lecture, or I won't have time to reply.
ReplyDeleteAnd please read all the preceding commentaries and replies before posting yours, so you don't repeat what has already been said.
** IMPORTANT: The mirror-neuron reading has been updated, The Rizzolatti & Destro reading has been substituted for the Bonini et al reading which was far too dense. My apologies.
Delete"Gradually abandoning the one-brain paradigm in favor of a multi-brain paradigm is of crucial importance to understanding the brain–behavior relationship in ecologically relevant conditions."
ReplyDeleteThis idea is interesting- and seems to raise some issues for reverse engineering the mind. If brain-brain interaction could play a causal role in cognitive processes, it would seem that an actual understanding (that is of another cognizer as such) is necessary for full cognition. Additionally, if the brain activity is based on empathetic perception of physical signs of emotion, it seems it would be necessary to grant a weak AI system visual capacity. That doesn't seem impossible, but very challenging (and interesting)!
T-testing already involves two cognizers.
DeleteWhat do you mean by a "weak AI system"?
What are "mirror neurons," what do they do, and how?
Please see the updated reading, Rizzolatti, G., & Destro, M. F. (2008). Mirror neurons. The Bonini 2022 reading was too confusing. I apologize.
DeleteBy weak AI system I was referring to a system that attempts to simulate human cognition, but not necessarily recreate it. I should have specified!
DeleteThis other article is very helpful- thank you for linking to it. Mirror neurons are neurons that activate when an action is performed and when it is observed- which could have implications for learning and social reasoning. They allow us to imitate and learn behaviour by activating the same pathways used in behaviour itself, allowing us to know the motor content of imitation without actually doing it. My original point in the comment above was that that could indicate the necessity of a community of cognizers, if many or most actions are understood sympathetically. However, I think I failed to realize that cognition is not realized via such a process, and that mirror neurons don't seem to provide a causal explanation for cognition itself.
Jacob, that's fine, but please also read the other comments and my replies.
DeleteMirror neurons were briefly mentioned during Friday’s lecture, to emphasize the difficulty in reverse-engineering humans’ cognitive capacities. Indeed, MNs allow us to observe others’ actions and imitate them, or flexibly use that information to generate socially appropriate actions. However, while the activity of MNs correlates with our neural capacities, they fall short of explaining how our brain actually ‘thinks’. In short, while MNs allow us to imitate others’ cognitive capacity, this does not explain how we are able to imitate this capacity.
ReplyDeleteSpot-on again, Amélie. MN activity is correlated with what we already knew human cognizers could do. Does discovering that MN activity is correlated with imitating others' actions explain how we do it? Have we reverse-engineered the capacity in discovering mirror neurons? (Think of what Fodor said about when/where vs. how/why...)
DeleteThe Rizzolatti and Destro paper was instrumental in helping me understand mirror neurons (thank you for updating the reading).
DeleteThe authors bring forward potential “functions” of mirror neurons: “action understanding” and “imitation” concerning understanding intention, empathy, and the evolution of language.
These are attempts to answer the how/ why of this neural structure but are only extremely partial answers (dealing with observational when/ where to a larger extent). The description does not lay out a comprehensive causal mechanism which would allow us to reverse-engineer human cognitive capacities.
Fodor would argue that, therefore, the MN are not worth studying as a whole. But I think the where and when questions are helpful in clinical settings and might potentially help us get closer slowly to the how and why (for example, language evolution).
Kayla Good summary. What is "action understanding"? And how are the capacities with which MN activity is correlated related to language?
DeleteI agree with Amélie about neural capacity vs. thinking.
DeleteOne of the lines from this article reads: “Taken together, these experiments suggest that feeling emotions is due to the activation of circuits that mediate the corresponding emotional responses.”
Is this a surprise? To me that seems a bit obvious. The phrase “Fear occurs in the amygdala,” might be true empirically, and we may be able to prove this with fMRI, but it still doesn’t tell us anything about the lived experience of fear, or any other emotional response. Maybe mirror neurons can tell us something about understanding why people behave a certain way in response to other people, such as the example with autistic children (which I’m hesitant about), but MNs aren’t the key to cognition.
Teegan, I had similar thoughts to you when I read the section on emotion and empathy. It does seem obvious that every human ability has a corresponding brain region or process that contributes to its functioning. It is still helpful to identify these regions (I imagine it is useful when assessing brain damage or other brain conditions), but simply acknowledging where something occurs does not reveal its process. Even examining these regions on a closer level will most likely not provide us with any more clarity. This highlights one of the main challenges in regards to reverse engineering- simply understanding the structure of something does not mean you understand how it truly works.
DeleteTeegan, MNs certainly don’t explain cognition, nor even help us reverse-engineer mirror capacities, but what is it that mirror capacities have in common?
DeleteAnger has two aspects: what you DO when you are angry and what you FEEL when you are angry. How is this related to mirror capacities?
(in response to the question "What is action understanding?" above)
DeleteAction understanding according to Rizzolatti is the process by which mirror neurons simulate (if this word can be used in this context) an action, copy an action's firing pattern in the brain, enabling "real experiential comprehension of the observed action." So to kid-sib: as we look at an action, our brain mimics the action so we can understand it's meaning.
This is correlated with language, as MN enable direct comprehension through gestures. The receiver can understand what a behaviour means (verbal or hand gestures for eg) through the MN imitation capacity- therefore a message has been communicated: passed from one person to another.
Hi Teegan,
DeleteI had a similar question regarding the relation between MNs and emotions/empathy. Thanks to MNs, we can empathize with what others "feel" by observing their motor output. But just like Searle's CRA, Searle can appear to be able to speak Chinese without actually understanding it by virtue of his verbal outputs; how does an empathizer know that he or she is empathizing with the same exact mental state of someone that's under their observation? If I understood correctly, rather than "reverse-engineering their feelings," the MNs are comparable to a program that provides output (to the ventral premotor cortex) based on its input (specific/effective visual stimuli). That said, the MN does not explain how we can empathize because we do so by interpreting the motor outputs of those whom we empathize with. In other words, the inputs (visual) and outputs (motor cortex activation) here are just symbols to be manipulated based on the program (MNs), which contains only syntax. Empathizing with someone means interpreting their motor output (semantics). However, the interpretation process is beyond the scope of the program (MNs) itself.
Kayla, mirror capacity comes from some sort of structural similarity between (the feeling of) DOING something and (the feeling of) SEEING it being done by someone else.
Delete(The feeling in brackets is there, but it’s a (the!) hard problem to explain the feeling’s causal role; roboticists may well succeed in producing the capacity to imitate in robots that do not feel.)
But you’re oversimplifying the relation between gesture and language. Pantomime, and any gestural communication, is not language (though it can set the stage for it). It is just imitation, sometimes being used to get someone to do the same thing you’re doing (and then it’s just induced imitation). Or it’s imitating, gesturally, what some object looks like (or vocally, what it sounds like), to get you to find it or think or know about it. That’s sort of like pointing, or charades, or “Pictionary.” But it’s still not language. Miming becomes language in a more complicated process (if it’s true) of two related transitions:
(1) The first transition is from iconic gestures (which resemble the thing they are imitating) to arbitrary gestures (which no longer resemble the thing they had resembled; they can then be used as symbols and names to refer to it). But naming things is no more language that imitating them.
(2) The second transition is from pantomime (acting out, based on similarity) to propositions, with subjects and predicates and truth-values (true or false). That’s what produces the power of T2, which can describe, define or explain anything. Without that it’s still just pantomime, even if it’s purposive (to draw attention to something, to “ask” someone about something, to warn someone about something).
But the sentence I am writing now is not doing any of those things. What is it doing?
Teegan, imitation, which is based on the similarity between seeing something and doing something, is not reverse-engineering! It’s based on the brain detecting and using the similarity between perception and production – between seeing and doing. And that includes the similarity between what I feel when I am stressed, and what I see you doing when you are stressed. So it’s not reverse-engineering (we still don’t know how our brain does it). But it is mind-reading. How? (And T-testing draws heavily on our mind-reading capacities.)
DeleteAnd communicating using mirror-capacities like miming is not language; but it has something in common with language, and perhaps helped us get to language. Language began with the transition from showing and telling. To tell someone something, you need words (grounded symbols) and propositions (Week 8b – by the way, be sure to follow the numbering for weeks 7 and 8, not the sequence because their orders got scrambled up, just as 4a and 4b did!).
Think about it.
Yucen, mind-reading with the help of mirror-capacity is not perfect, but it’s very powerful. You don’t always understand what someone else is thinking or feeling, but amazingly often, you do. (And what about language?)
As for Searle Searle, compare Searle, when he is squiggling in “Chinese” with Kayla when she (or, for that matter Searle) is speaking in English in the real T3 world (not just the word-land of T2). Ask Kayla to point to an apple. Then ask Searle, in Chinese, to point to an apple.
That said, there is some mind-reading in language speaking and understanding, and mirror-capacity does play a role in that: What is it?
(Don’t forget that reverse-engineering is done by cogsci, not by normal people, when they are showing or telling one another something, with the help of mind-reading using their mirror-capacities.)
"That said, there is some mind-reading in language speaking and understanding, and mirror-capacity does play a role in that: What is it?"
DeleteMirror capacity refers to the connection between sensory perception and motor production. To my understanding, the mirror capacity equates mirroring perceptual and motor patterns. The mirror neurons are responsible for the neural pattern similarities between "what I feel when I am stressed" and "what I see you doing when you are stressed." Similarly, when it comes to mind-reading in language speaking and understanding, mirror neuron capacity makes it so that no arbitrary symbols are needed for mutual understanding since the "understanding" is already inherent in the neural organizations of the two individuals.
This paper relates back to Fodor’s point in the last reading. Before discovering mirror neurons, we already knew that we could recognize and imitate movements of others. After discovering mirror neurons through neuroimaging, we now know it scientifically. This discovery has clinical benefits, such as for autism and developmental disorders, and psychiatric and neurological diseases. However, as Fodor expressed, it produces no benefit in terms of understanding how and why we are able to do the things that we can do. Knowing what mirror neurons are does not bring us any closer to explaining how they work or to reverse-engineering T3 capacity.
ReplyDeleteGood critique, but what about mirror-capacity itself (apart from its neural correlates)? What do imitation (manual and vocal), the "understanding" of movement and intention (not "intentionality"), the understanding (and production) of language, mind-reading and empathy have in common?
DeleteThese are all functions which require involvement of the motor system. That is what MNs do: they provide motor copies of observed actions, which facilitates understanding. For example, the article describes the theory that we originally communicated solely through gestures, then sounds became associated with these gestures and this evolved into communication via language. MNs, originally responding just to the gestures, evolved to respond to the associated sounds and then to language. Thus, the motor system was fundamental in the development of the human capacity of language.
DeleteThe functional roles of MNs, like understanding and production of language, are grounded in activation of the motor system. We can infer then that understanding requires the motor system. This substantiates the claim that T3 capacities are necessary to pass T2; without sensorimotor capacities, MNs cannot foster understanding.
Josie, on “association” see reply to Sophie below.
DeleteIn the reading, the idea of multimodality is discussed as, even at the neural level, “constituting a necessary condition for the sensory systems and the motor system to work together.” This touches on the concept of embodied cognition, which emphasizes how the body and mind work in tandem to shape cognition; the brain, body, and environment are interconnected. Embodied cognition supports the notion that T3 is necessary to pass T2, since sensorimotor capacity is a necessary aspect of cognition. “Grounding multimodality on motor and sensory neurons emphasizes the relevance of our bodily experience of the world over abstract and formal computational logic.” Here, the reading specifies that cognition is not computation alone, it requires bodily experience.
ReplyDeleteCorrect, but what is a mirror capacity?
DeleteAccording to the updated reading, “mirror capacity” seems to be the capacity for imitation, action-understanding, intention-understanding empathy and language perception and production. This capacity was present before mirror neurons were discovered and they seem to play a major role in our ability to produce these capacities. The “why” of mirroring capacity seems to have many potential explanation both genetic and cultural but it’s the “how” that’s less obvious and might give us more insight
DeleteWhat is "mirror capacity"? Action-understanding? Intention-understanding? empathy? and how is mirror-capacity related to language?
DeleteMirror capacity is the capacity for imitation, action-understanding, intention-understanding, empathy, and language perception and production. The article discusses how similar brain activation occurs when we are doing an action and when we are observing someone else doing them. This is involved in action-understanding because MNs can differentiate between actions based on goals and reasons (e.g. grasping an object for eating vs placing). The context matters. Similarly, intention-understanding is enabled through MNs differentiating between the deliberateness behind the movement. Thus, mirror capacity depends on the shape of the movement and the shape of the context of the movement (intention/goal). Mirror capacity is not only good for movement but also empathy, which involves mind-reading: MNs enable our sense of what people are feeling based on what they are doing. Language is another mirror capacity that involves more than just mindless imitation (unlike pantomime). Rather, language necessitates meaning/semantics and, particularly, the mirroring of meaning. By engaging in language perception and production, we show understanding of the meaning behind the sentences both when we produce them and when we perceive them. These mirror capacities all involve a connection between sensory perception and motor production, and they indicate understanding.
DeleteIn this paper, mirror neurons are described as a revolutionary discovery that will “build bridges between neuroscience and humanities”. In particular, the evolving research around ‘other-selective’ neurons has revived interest in the mirror mechanism: the evidence showing that they are found in the frontal motor area, cohabitating with neurons selective for the self, sparks hope for a refined theory of single-neuron mechanistic explanation of emotional social processes, an understanding of the neural correlates underlying this mirroring capacity. Thinking back to Fodor’s argument, we can doubt here the relevance of focusing on reverse-engineering our ability to mirror actions through identifying the specific places in our brain which ‘light up’ when we do, or even more precisely specific neurons that are activated when we do. Indeed, this does help us in our quest to find out how and why organisms can do (mirror others) the way that they do.
ReplyDeleteMathilda: I couldn't follow what was revolutionary and what was explaining what: could you explain it to kid-sib, please? And what is it that has suddenly made reverse-engineering irrelevant: Do we know how mirror neurons produce mirror capacity? We need a causal mechanism, not just a mysterious correlation!
DeleteWhat I tried to say in my comment is that the claim this paper makes, which is that mirror neurons (and other-selective neurons) are a revolutionary discovery which will enable us to explain emotional processes, actually seems irrelevant when thinking back to Fodor's article. Researching motor neurons and the neural substrate of the mirroring mechanism does not allow us to reverse-engineer organisms' ability to produce mirror capacity (the how). And reverse-engineering is what cognitive science aims to do. Therefore, cognitive science should aim to focus on other ways of explaining our abilities than studying the correlation between brain activity and the ability to do what we can do.
DeleteAn interesting section of this dense paper was talk about the emotional display of others and how that is orchestrated by a set of deep brain regions connected to the somatomotor cortical circuit. This emphasizes that our emotions are primitive and crucial in society to coordinate behavior based on emotion. In the realm of affective neuroscience, I have always wanted to understand the neuroscience of empathy, and why some people are so good at it and others so lacking. Could these mirror neurons be involved? With emotion being at the heart of cognition, it would be useful to understand whether these mirror neurons in charge of recognizing emotion are simple, singular, and mapped in a one-on-one fashion, or rather are large, interconnected networks relaying to different sensory regions of the brain. Do our emotions spring solely from our own minds or are they fundamentally contingent on the emotions of others?
ReplyDeleteHow is emotion at the "heart" of cognition? And if activity in some neurons is correlated with empathy, what have we learned about how the brain empathizes?
DeleteWe have yet to learn how the brain empathizes. Since we can only observe if activity in neurons correlates with empathy (or any other emotion), we’re stuck looking at the hardware, which teaches us nothing about how empathy arises in us. While the clinical benefits of MN are impressive, especially when it comes to aiding rehabilitation, I think this paper only reinforced the idea that computation isn’t enough by jumping from one framework of MN mechanisms to the next. Indeed, because the scientific jargon is so dense, it just left me confused and wondering: “What is this really explaining?” We’re getting explanations that serve the easy problem, but that’s it.
DeleteRosalie, I've substituted another reading. Your summary is right, and the Bonini 2022 reading was indeed confusing. I apologize. I hope Rizzolatti & Destro will be clearer.
DeleteI now read the second paper which makes a lot more sense. Other than the discovery of motor neurons highlighting the fact that our capacity to imitate and understand actions and empathize and produce language, we still have no clue why the act of looking at another person doing these things or feeling these things will activate these certain neurons. I’m asking myself though if there is ever any way we’ll be able to pinpoint how and why our brain can do what it does. And why is it so essential to know how mirror neurons are activated in the brain rather than knowing that they do? I hope I'm not repeating myself too many times, but I still think there is value to knowing that empathy and intention-understanding spring from these mirror neurons, especially in the study of certain emotional disorders like autism, as mentioned in the article.
DeleteTess there's no doubt that clinical neuroscience can be clinically useful. But how does the discovery of MNs help cosci reverse-engineer even just imitation? Have a look at the robotic article: It takes note of the fact that MNs have been discovered to exist, but they do not give a clue of how to give robots the capacity to imitate.
DeleteHonestly, I found this paper boring and dense. They listed all these ways that mirror neurons are involved in so many different cognitive processes from empathy to motor imitation. Specifically, mirror neurons were discovered, then promptly deemed revolutionary evidence of an instance of 1-1 mapping for a specific behaviour to a specific brain area. However, this paper just lists all these ways that this is an oversimplification. They even go so far as to state that imitation, as thought to be controlled by mirror neurons, likely involves a complex system, including multiple brain areas and networks. This seems to be a trend neuroscience, especially those involved in more complex cognitive and affective abilities. But overall, none of this provides real insight into how/why we really think and feel. It's a bit discouraging, and maybe it is just the process of science, the more we know, the more we know we got wrong. Similar to Fodor, I struggle to see how neuroscience is going to solve the Easy Problem, let alone the Hard Problem without first acknowledging that it is asking the wrong question with functional mapping. I guess this is where cogsci comes in, but again very curious to see what methods may be more effective at approaching this problem going forward.
ReplyDeleteSophie, you're right, the Bononi 2022 reading was too confusing. I've now substituted Rizzolatti, G., & Destro, M. F. (2008). Mirror neurons. which is much simpler and clearer. I apologize.
DeleteAfter reading the second paper, the role of MN became more clear. I understood that not only MN were important for understanding, but also for intention understanding, among other functions such as language and empathy. The experiment that proved the hypothesis for intention understanding is the one where volunteers "were presented with hand actions without a context and hand actions executed in contexts that allowed them to understand the intention of the action agent. The main result of the study was the demonstration that actions embedded in contexts yielded selective activation of the mirror neuron system." However, even though that reading is clearer than the other one, it still shows that knowing about MN doesn't add much to cognitive science since we don't know how it happens.
DeleteCharlene, good points. How about language? In what way(s) is it a "mirror" capacity?
DeleteThanks for providing the other reading! This one was definitely more straightforward. To answer your question, mirror neurons seem to be what allow people to understand language without any “cognitive mediation”. They discuss a theory in which language began with gestures, then sounds became associated with those gestures. Mirror neurons, which foster understanding of gestures by activating the observers motor system, also started responding to the ‘sounds’ that were associated with said gestures. This then entrained them to make implicit associations from objects to words. For example, if I were to say “I am holding a Banana” your mirror neurons would activate areas in your brain in the same way they would if you actually saw me holding said banana. This is pretty cool, and I wonder how it could be useful in solving the symbol grounding problem.
DeleteSophie, Rizzolatti does not mean language begins in children with gestures. He means it evolved in our ancestors, 50-150 thousand years ago. (1) First gestures (pointing, imitating), which is not language, though it is communication. (2) Then gestural language (see week 8b). And (3) language then migrated from the gestural modality to the vocal modality because of its advantages over the gestural (what are they?). There were radical genetic and brain changes with (2) and (3); probably also with purposive communication (1). In contrast, writing was learned, not evolved, and as far as I know, entailed no genetic or brain changes. (But I bet you can find some things in google scholar that say otherwise,)
DeleteYes, this is related to symbol grounding (Week 5), but does not involve “associating” gestures with sounds. Symbol grounding starts before symbols and before symbol grounding; it is connected with category learning (Week 6), which preceded language, and is part of the cognitive capacity of many species that do not have language. (What is language? How might it be connected to Mirror Capacity?)
My instinct is to say that language is a system of communication consisting of rules understood by all users of that language in the same way. This is why merely gesturing (1) cannot be considered a language; because there are no universal rules in place. Pointing or waving may be intended to have one meaning by the initiator, but is interpreted a different way by the receiver. Those gestures in the context of a language, like sign language, would have the same meaning to both initiator and recipient. If mirror capacity is about the capacity to perceive others’ intentions and emotions through their actions, then mirror capacity would be integral to the understanding of language, as agreement on the universal rules is necessary.
DeleteJulia, sign languages are not pantomime or imitation. They are full-blown languages, in which you can say anything you can say in English (or any other language). You cannot mime everything you can say in English! Try it on a few sentences in this thread. You’ll soon give up and realize why words had to be arbitrary, not resembling the shape of what they refer to.
DeleteAnd yet mirror-capacities may still play a role in languages – signed, spoken and written. How?
Though the exact role mirror neurons play in memetic capacity, and how they do it is unclear, they indicate that our capacity for understanding intentions is linked to mirroring. As is said in the article, through the mirror mechanism “actions done by one individual become messages that are understood by an observer without any cognitive mediation." It would make sense that language evolved from gesturing, as it is an efficient way of communicating intention and goals, essential to the types of cooperation and competition we engage in socially. From my understanding of mirror neurons, they are located in the premotor cortex which deals especially with planning and initiating motor actions relating to objects in extra-personal space. This is a limitation of a mirror-motor mechanism: much of what we want to talk about lies beyond immediate extra-personal space in the past, the future, in hypotheticals and abstraction. To talk about these things requires something disconnected from the immediate physical environment, another way to communicate meaning then by intuitive mirror capability.
DeleteConsidering this last question of how mirror-capacities may yet play a role I language, I would suspect much of it lies in making sense of the gestures and facial expression that accompany words.
DeleteHadrien, you’re right that more is needed for human (and nonhuman) cognition than mirror-capacities. An example is learning; another, unique to humans, is language.
DeleteSee other replies about the difference between language understanding (and production) and senosrimotor know-how. Understanding what a proposition (“the cat is on the mat”) means is different from seeing (and “understanding”) that the cat is on the mat.
Another reason gesture is a better modality for miming is that it can imitate a lot more than vocalization can. So it’s a more likely place for language to start. (But once started, vocalization quickly becomes much more suitable than gesturing. How?)
Once propositional capacity (and motivation) begins, out-of-sight things can be talked about regardless of whether language is gestural or vocal. (What is a proposition?)
It is true that using vision to communicate about objects we mainly perceive visually is a sensible starting point. However, any gesture-based language requires communicators to be within sight of each other; increasingly detailed sequences of gestures not only requires unobstructed sight, but also enough proximity between communicators for detailed perception. Gesturing limits effective communication in darkness, and in daylight demands precious visual attention better devoted to other important things, like watching one’s step. In many, many circumstances, simply having to be within earshot to communicate has a staggering advantage over sight-dependent communication.
DeleteAs for what a proposition is: a declarative statement in some language that describes something (I'd say objects or events in general), and (usually) can be judged as true or false relative to what one means to describe with it.
DeleteHadrien, good replies.
DeleteI think that language is a mirror capacity because the it is said in the readings that language typically evolved from the basic neural mechanism that is the mirror system. Moreover, we talked about the mirror neurons being involved with actions in a certain context, and language can be gestures. We can for example think of sign languages that are hand gestures produced with the intention to say something. Therefore, we can say that the discovery of mirror neurons provides a strong support for the gestural theory of the origin of speech.
DeleteI agree with my colleagues that research on mirror neurons so far provides little insight (cognitively speaking) as to on how a mirror capacity works. It seems to be an entirely accidental fact of the architecture of the brain that the same population of neurons discharges both when encoding other-related information (e.g., observing a goal-directed action) and associated self-related processes (e.g., executing a similar action oneself). It is perfectly possible, at least in principle, that the ability to mimic another's behavior could be implemented in a neural network with a different design. After all, what does it matter what neurons encode what information where, so long as the same inputs go in and the same outputs go out? Perhaps there is something to be said here about the efficiency of the mirror neuron system. If it turns out to be the most cognitively efficient way to implement an imitation capacity, this would at the very least explain why this system stuck around in evolution.
ReplyDeleteFodor would be quite right to point out that the "discovery" of "mirror neurons" -- whose activity is correlated with imitation, action-understanding, intention-understanding, empathy, and language perception and production – does not explain how “mirror neurons” produce our capacity for imitation, action-understanding, intention-understanding, empathy, and language perception and production.
DeleteWe already knew we had the capacity for imitation, action-understanding, intention-understanding, empathy, and language perception and production before mirror neurons were ever discovered. Let’s call those capacities “mirror capacities.” The correlations with neuronal activity and connections so far explain almost nothing (although their geography does provide some support for the hypothesis that language originally started in the visual/gestural modality rather than the auditory/vocal modality to which it later migrated; we will talk more about that in Weeks 8 and 9).
What mirror capacities have in common is some sort of analogy and connection between sensory perception and motor production. This perception/production mirroring, too, was already inherent in our known capacity for imitation, action-understanding, intention-understanding, empathy, and language perception and production, but the neuronal correlations made it more evident.
But to appreciate how little reverse-engineering the discovery of MNs inspired, you just have to look at the robotic research – both reverse and forward engineering -- on giving robots the capacity to imitate one another’s movements (e.g., in the optional readings, Bandera, J. P., Marfil, R., Molina-Tanco, L., Rodriguez, J. A., Bandera, A., & Sandoval, F. (2007). Robot learning by active imitation. INTECH Open Access Publisher).
This gives you a good intuitive index of whether a new discovery in neuroscience has helped move us closer to solving the easy problem: Does it help roboticists produce a robotic capacity that they were unable to produce before the new neuroscientific discovery?
The commonality between the mirror capacities does indeed seem to be a “sort of analogy and connection between sensory perception and motor production”. I agree with this identification but also believe there is a bigger picture pertaining to its relevance in reverse engineering human capacities. The article demonstrates that neurobiologically there is similar activation in specific regions of our brain when we not only do/feel certain things (grasping, emotions etc..) but also observe them. But what I found to be a more profound application, towards our goal of reverse engineering human capacity, is the idea of ‘goal understanding’. This idea is supported by the Fogassi et al. experiments where neurons were seen to selectively discharge during observation of motor acts that were embedded in a given action “(e.g., grasping-for-eating but not grasping-for-placing)”. Although the finding is just attributed to intention-understanding, the more profound application here is that if we are to produce a learning robot representative of our human capacities, it mustn’t just mimic or imitate our action or emotion intentions, it must know the goals & reason of our actions or emotions. I know this article doesn’t show this directly, but it does lead to further experiments and studies that do (kinda). An example is this study in which monkeys used normal or inverted pliers to grasp an object. The study found the same corresponding brain regions lit up for either action of closing hand (normal pliers) or opening your hand (inverted pliers)- suggesting a goal that must be understood. Such studies lead us (or at least leads me) to postulate whether our learning robot needs to infer our goals/intentions to mediate learning- which isn’t inherently required for T3.
DeleteSepand very good points! (And the MN studies of the kind you mention can be done, non-invasively, with humans, instead of hurting monkeys.)
DeleteThis paper reinforced the idea that humans have shared representational frameworks (mostly) grounded in sensorimotor experience that allows us to do all of the things we can do (cognition). Thus, while neuroscience can clinically inform treatment for disease, disorders, and injury of the central nervous system, the mechanisms that underlie these individualized deficits (as well as the individual variability itself) are irrelevant when it comes to understanding how to reverse-engineer the brain’s capacity as we share the same general abilities.
ReplyDeleteDarcy, what are the "mirror capacities"? What do they have in common? (I'm just asking for a description of the capacities, and in what sense they are each mirror capacities. I'm not asking for their how/why explanation; there isn't one yet.)
DeleteFrom what I understood in the Rizzolatti reading, these ‘mirror capacities’ represent a “mechanism that maps to a symbolic description of actions carried out in the higher order visual areas onto their motor counterpart.” In other words, it acts as a connection between the interpretations of our visual input that is ‘mirrored’ in the motor areas. Action understanding, intention understanding, imitation and empathy are all examples of how interpreted visual input maps onto motor areas, and can get exhibited through behaviour.
DeleteMirror capacity is the capacity to perceive the intention and feeling behind what someone else is doing. Mirror capacities include imitation, action-understanding, intention-understanding, empathy, and language perception and production. All (5) mirror capacities require a connection between sensory perception and motor production. They also require an ability for goal understanding (knowing the goal and reason of actions and emotions).
DeleteThis article made me think of what Fodor said in the previous reading: “The empiricists won that battle, of course; but my guess is they will lose the war”. Fodor was a bit right because the article about mirror neurons (MN) show that different clusters of neutrons in different areas of the brain get activated in reaction to different emotions or actions performed by others (compared to the empiricists who believe that). These MN allow us to imitate movements, recognize vocalizations and produce them. However, because we cannot explain how we can do that, I don’t think that what the article is explaining about the MN system is a “revolutionary discovery” since it doesn’t add much to cognitive science. From what we learned in class, the goal of cognitive science is to reverse-engineer the human brain, but since we don’t really know how the MN system works, it would be impossible right now to build a system with the same mirror abilities. However, let’s mention that, as said in the article, the discovery of MN led to a better understanding of autism or neuro-degenarative disorders, etc. There is one thing that I was wondering, and it’s whether or not the MN are more reactive when the other who is observed is a family member or someone close to the observer.
ReplyDeleteCharlene, correlates can be relevant for diagnosis and treatment; and your last question would certainly be feasible as a senior thesis. But suppose MNs are more active with family than strangers: what next?
DeleteI’m not sure if this would be a stretch, given the information in the article. However, if motor neurons are linked with intention-understanding as well as the (more simple) action understanding, would it not be expected for mirror neurons to be more active for family members than strangers, as we would have more schemas to understanding their actions and intentions. Since motor neurons link the observer with the action’s agent through a means of social interaction, if those social interactions are more readily available (or common), such as in the case of family members, I would expect stronger activation of these MNs, or at least a faster activation of them.
DeleteKarina, perhaps.
DeleteAs Karina said, it is easier to imagine a stronger activation of the mirror neurons in reaction to a family member’s action rather than the opposite. Since we would already have a certain degree of what to expect from someone we know, we can imagine our understanding time reaction to be a bit faster. Also, from an evolutionary point of view, I can imagine that is was more useful for survival to have fast reacting mirror neurons to a related other’s action in case of danger for example, compared to a stranger who wasn’t in the same clan and didn’t know about the situation, context and dangers threatening the members of a same clan / family / community.
DeleteThis discussion on how MN may react differently towards family/strangers made me think about how in the article, the monkey’s MNs become active from observing a human’s actions, meaning these MNs react cross species. It would be interesting to consider to what extent similarity is necessary in order to activate these MN (perhaps by seeing if these mirror neurons would activate when shown the hands of possums or chameleons for example). If these MN only require visual observations in order to be activated, one could put together clips of different motor acts of different species to explore how MN activation levels differ. It would also be interesting to see how the monkey would react to viewing a motor action that is physically impossible for them to imitate (example: viewing a simulation of another monkey overextending its joints).
DeleteWhile this paper gives a clear understanding of the functions and features of mirror neurons, I believe it still does not give a clear answer to the “how” question about the brain. Mirror neurons are activated during observing, imitating, etc., but how do they do this? They might be supportive in explaining some of our actions, but it is not a sufficient answer to a causal mechanism and how and why we can do the things that we can do. For example, why are they activated? How does our brain know that it should activate mirror neurons, in order to imitate an action?
ReplyDeleteAfter reading this paper on mirror neurons, I was left thinking about how it would apply to robots, especially being able to implement it in T3 robots. I think that after reading this paper we could say that MNs are potentially very important to cognition seeing as they add complexity to the reverse engineering of the brain.
ReplyDeleteThis paper shows how the brain itself is not an isolated system but rather one that needs constant interaction with our environment/surroundings. In this case, it is possible to say that even if we are able to understand the brain on its own, we would still be missing information gained from the constant interactions with the environment to fully understand cognition.
Now that we know MNs are consequences of learning mechanisms, the question would then be how, or if, we can implement such MN mechanisms computationally into our T3 robots.
Melis, what is it that the different Mirror capacities have in common?
DeleteAccording to the reading, the functional role of mirror neurons are hypothesized to be action understanding, imitation, intention understanding, empathy + basic neural mechanism from which language evolved. All of these hypothetical functional roles involve interaction from one cogniser to another.
DeleteI find it interesting to think about how in order to achieve action understanding it is necessary to employ the motor system. As Melis said, one can apply this to robots, notably at T3 level since, similarly to mirror neurons, action understanding/imitation go hand in hand with interaction of the observed environment.
I also associated the explanation as to how visual perception provides description of movements but does not provide information on the action to the analogy you gave us involving the Chinese characters.
Ariane, but how can you imitate the movement you see without perceiving the movement? Mirror-capacity is mirroring perceptual and motor patterns; you need both.
Delete(I didn’t understand your last paragraph.)
Looking back at my last paragraph after having further learned this course, I believe what I was trying to say here was an underdeveloped version of what we have just covered regarding symbols. The reading explained that description of a movement is not enough to provide information on the action... similarly to when it comes to symbols, they are interpretable as having meanings, but their shape is arbitrary in relation to their meanings. At least that's the connection I made, perhaps the two are not at all related!
DeleteI want to make a quick parallel between this paper and Turing's idea of creating a "child-like" mind.
ReplyDeleteWhen Turing talked about creating a child-like learning machine, he believed that this would be much simpler to develop Artificial Intelligence than making an already "matured" mind.
In this paper, the authors present the idea that language evolved from our ancestors' mirror neurons when communicating with gestures, followed by sounds and "speech."
Would it then be possible to create such an algorithm to simulate Mirror Neurons? If you can simulate Mirror Neurons, you could technically replicate a very primitive form of our ancestors' brain and use that as a starting point to learn more about how our brains evolved. This could be a stepping-stone in also creating the "child-like" mind that Turing talked about.
And sure, we might not understand how intelligence develops from mirror neurons, but as a Lilliputian, I believe it's a step in the right direction.
Alexei, language may well have evolved from gestural communication (pantomime, which depends on imitation, which depends on mirror-capacity), but pantomime is not language. And vocal language depends on mirror-capacity too. So what is language? (See other replies, including the two transitions – iconic to arbitrary and pantomime to proposition, show to tell – but telling does not have to be vocal. Sign languages have all possible subject/predicate propositions too; and the remaining iconicity of some of their gestures is as irrelevant as hardware is to computation.)
DeleteBut here’s a challenge worth an A: Why might gestural imitation be a more promising way for language to start than vocal imitation?
But forget about “replicating” mirror-capacities computationally. You can computer-model them (just as you can model anything else, because of the Strong C-TT), but squiggled mirror-capacities cannot even move or see (or freeze, or overheat). They are just squiggles that are interpretable (by us) as moving, or seeing, or freezing or melting (if the simulating algorithm is close enough.
I think this concept of the "child-like" mind is noteworthy in the context of mirror neurons. I believe that reverse-engineering a more rudimentary mind with the capacity to grow and learn as a human mind does sounds more promising than reverse-engineering an adult mind since a younger mind can, using its mirror mechanism, learn from social interactions, adapt accordingly, and develop a unique personality. While a "personality" may seem superfluous in our cognitive pursuit, I cannot imagine that human-like cognitive capacities can occur in an entity without one...
DeletePolly, yes, and people have real-time histories, which would be easier for the T3 to learn than to have built-in in advance. It’s less a childhood that’s needed than the capacity to learn.
DeleteI might take a stab at the question, "Why might gestural imitation be a more promising way for language to start than vocal imitation?" For me, it seems that with gestural imitation, the gestures which are the symbols transmitted during communication are more grounded in real-world referents that are visible or can be visually indicated in one's surroundings. Since the formal structure of our linguistic communication seems arbitrary with respect to its meaning, it makes sense for there to have been an initial mapping between signs more explicitly grounded in what they signify, from which more arbitrary linguistic signs could have developed.
DeleteZahur, that's one of the two main advantages of gesture over vocal imitation: it has incomparably more scope for mapping mime onto its sensorimotor referents than vocal imitation does.
DeleteThen there's the gradual transition from iconic to arbitrary shape, as gestures for the same thing are shared and become conventions in a community.
Next, the fact that these increasingly arbitrary symbols are sequential; then the sequential order, too, can become conventionalized or standardized by order-conventions, like action – doer – effect (the beginnings of syntax).
But the most important feature for a transition from pantomiming to propositionality once miming has become a formal series of arbitrary referent-names is that of {Subject/Predicate}, along with the all-important “propositional attitude” that [it is true that] “{cat / is on mat}.
There are still some mysteries about that transition: Was the ”propositional attitude” first a learned practice or motivation that then became a fundamental innate feature of language in our genes and brains? Or did it already start out as a genetic mutation? (More on this in Weeks 7 and 8.)
When contextualizing mirror neurons into the material we have covered thus far in class, it prompted me to consider the similarities between the actions of MNs and Searle's Chinese Room argument. In particular, how as
ReplyDeleteSearle was simply performing computations when "speaking Chinese", the reflections of the mirror neurons are also simply computation of squiggles (higher order visual information) into squaggles (non-discharging motor cortex output). In both cases though, they lead to a functional output that works to serve some sort of purpose.
Emma, what is computation? The shape of a symbol is arbitrary, whereas the brain has many analog (iconic) properties, where the “shape” of the neural activity resembles something outside the brain (for example the shape of the tonotopic higher visual areas, which preserve some of the shape of an image projected on the retina; or the intensity or frequency of a sound in the auditory system). Shape can be encoded in symbolic algorithms too, but it needn’t be, and sometimes it is much less efficient than staying in analog. It will be interesting to see whether ordinary robotics (doing engineering rather than cogsci reverse-engineering) will find it more productive to implement mirror-capacities analogically rather than computationally. (And. before anyone asks, no, so-called “analog computation,” as by a sundial, is not computation [symbol-manipulation, squiggling].)
DeleteFrom what have been said about the mirror capacities in the replies, that the finding of the correlates of the MNs with the capacity of imitation, understanding of movement and intention, language perception and production, emotions and empathy, which we can now call all of them as mirror capacities since they all correlated with the MVs. When we imitate others we observe and repeat their actions, our brains can convert input sensory information into motor commands, comprehension, as well as emotions. So that we can imitate, understand, and feel. One of the differences of these capacities is we can imitate movements by only converting inputs into motor commands without the comprehension or the emotion part, where we can just repeat aimlessly, like robots. What is interesting is the fact that we can repeat movements without knowing how (unconscious knowledge). So we have the capacities involves unconscious knowledge and capacities involve conscious knowledge, too. It seems difficult to me reverse engineering either of these two kinds of capacities since we know the mechanisms of neither of them.
ReplyDeleteCynthia, I think you are over-interpreting and sometimes misinterpreting the MN correlations. They seem to mark a kind of congruence or invertibility involving perception/production pattern similarities or connections in the brain, some of them innate, some of them perhaps learned. We can, for example, imitate vocally (we hear ba and we can say ba) because our brain detects or creates a similarity between the heard sound pattern and our produced movement pattern.
DeleteThat’s the basic phenomenon. With purposive movements it is not just the shape of the movement that is perceived but, from the context, and background knowledge, also the goal of the movement.
With emotion, it’s not just empathy; herbivores can perceive the aggressive intent of their carnivorous predators although they do not themselves ever produce or feel that kind of aggression.
And the role of mirror capacity in producing and understanding language is even more subtle (if it exists), involving the words of others and our own experience (grounding) in the things the words refer to.
Mirror neurons are neurons that fire both when we are executing and observing a task. In other words, these neurons allow us to simulate the behaviors that we see around us. This article reflects Fodor’s idea on studying the brain. It is a good thing to have discovered mirror neurons and to be able to put a name on this particular part of the brain. However, this does not help us in knowing and understanding how we do it.
ReplyDeleteI want to reflect on the ideas that these two texts convey. I understand that it is not through studying fMRI scans and dissecting the brain into different regions that we will understand how intelligence works and how one comes to think. Nevertheless, I cannot seem to envision a different working method for neuroscientists. We have the brain; we have tools to study it and we have a scientific discipline explicitly created to obtain “knowledge of the nerves” (Neuroscience). I am in favor of the idea that a cooperation between scientific fields must be met in order to understand how one thinks, but you cannot blame the neuroscientists for working in their field of study, with their methods. What else could they be doing?
Hi, I agree with you but I think the criticism is mostly from cognitive science's point of view. The criticism is about neuroimaging but we can extend the criticism in the case of the approaches to study cognition. We can think of Fodor's criticism to neuroimaging as a prompt to think about what is the difference of cognition and the brain. The brain is made of neurons, cognition is cognitive capacities. They are not the same thing by nature to study.
DeleteMirror-capacity is not necessarily simulation; it could be sensory/motor pattern similarity detection (or creation, if learned).
DeleteNeuroscience is important for medicine and it may also give ideas to cogscientists on how to reverse-engineer cognitive capacity. But reverse-engineering the brain is very different from reverse-engineering the heart where there is an observable relation between structure and function.
And don’t forget the potential power and usefulness of computer-modeling for reverse-engineering.
This reading explains MNs and many of the hypotheses of their purpose in cognitive processes. However, as many posts before me have mentioned, it does not give much of an indication as to how the MN system actually works. Despite this, it is helpful in eliminating a few hypotheses for how we understand purpose and intention, such as having a separate 'dictionary' for interpreting visual input of others' actions. It also helps tie the ability to interpret others' actions as we would interpret our own to the initial construction of language as well as observational learning.
ReplyDeleteAnd it suggests to computer-modellers that there may be deeper similarities among different mirror capacities.
Delete
ReplyDeleteAccording to the reading and previous posts, the mirror capacities indicate "they represent a mechanism that maps the pictorial description of actions carried out in the higher order visual areas onto their motor counterpart." I find it interesting how the connection between sensory perception and motor production seems to be the underlying mechanism for various functions. However, even though the reading suggested MN enables this mechanism which supports all the mirror capacities, we still don't know how this mechanism works.
See above reply.
DeleteAs mentioned by many students, mirror neurons play the distinct role of observing and imitating similar or identical motor acts. However, to add to why this becomes a challenge with reverse engineering, it is not as easy to build something that can truly understand what is going on and react in the same way. Specifically, as an example, for these neurons to fire, would imply that they "understand" that a portrayed action is similar to/the same as what was portrayed. Therefore, this becomes a difficult aspect.
ReplyDeleteYou would need to know whether what you are building, had truly understood the action for the neurons to fire. In other words, you can easily design something that can act the same way, however, does it really understand why its mirror neurons fired or was it just programmed/built to do so?
Also, I'm unsure if this is a proper correlation. But I guess it can be of somewhat similar analogy where, you can make a machine learn a language, but did it REALLY understand the language?
I think we’re getting a little carried away with the notion of understanding. Understanding mostly means understanding language: knowing what words and sentences mean. That understanding is what Searle did not have, for Chinese. Language is not just something you know how to DO with your mouth, it’s not just a motor action, a movement. But imitation – seeing and imitating someone else’s movement is just movement – a capacity to do something, “KNOW-HOW”, « savoir-faire ». But, as we learned in Week 1 (1st grade school teacher), most of the things we are able to DO, we don’t really know how we do: our brains do them for us, and hand them to us on a platter, with no idea how we did it.
DeleteSo we know how to imitate. There’s nothing else to “understand” in imitation. We just do it, because we can – no questions asked. And we’ve seen that learning that MNs are active when we imitate does not explain how our brain does it. It’s only after cogsci successfully reverse-engineers the mechanism underlying that mirror know-how, and explains it to us, that we will understand that know-how.
In speaking and replying to language, however, there is understanding involved: understanding what words and sentences mean – especially what they refer to. “Apples” are things in the world that the word refers to. We know what they lok like and what to do with them. This is where mirror-capacities touch understanding, because we understand what words mean when we say them, and we understand what they mean when we hear them. So there’s some sort of mirroring going on there, but not just the mirroring of movement, like when we imitate the sound “ba”: the mirroring of meaning.
(But this is still very vague, and new, and maybe it will not turn out to bring new insights into reverse-engineering language production and perception. The “understanding” of action goals and the mind-reading of emotions may give some clues).
I completely agree with the association of meaning in terms of language (speaking and replying) and therefore understanding being involved in the evolution of language. However, isn’t this meaning later gained after we have learned to mirror the sounds? I am thinking of a child who learns to reproduce noises such as “ba” and “da” and then learns to group them together, without necessarily applying meaning to them? Is this not still constituted as language, and would be linked to mirror neurons without the semantic knowledge being necessary? I agree that this doesn’t explain the complexity of speech, but would believe this serves as a stepping stone to learning language.
DeleteKarina, see reply to Zahur above about pantomime and propositionality. If that scenario is right, it favors gesture of speech as the starting point for language. Once launched, language can migrate to the vocal medium from the gestural one, but not before.
DeleteUnfortunately, I read the original reading before you sent the updated one. I will read it too, but for now I will comment on the original reading because I want to post as early as possible.
ReplyDeleteWhat interested me most about the reading was the large influence that mirror neurons have on explaining human-human interaction. It seems the data somewhat suggest that our interactions are carried out by physiological mechanisms over which we do not have control. In regards to the predictive ability of MNs, the author suggests that "interbrain synchronies guide social interactions by means of underlying neural machinery in which self-related neurons" in one subject's brain cause action and subsequent activation in the other subject's brain. This makes me wonder to what degree our person to person interactions are cognitive or guided by invisible underlying mechanisms. I understand that this does not explaining exactly HOW we perform such interactions, but it is quite surprising in considering a phenomenon such as human interaction that is believed to be so personal and emotional.
I think you will find in the latest reading that direct link of human to human interactions explained by the various capacities described. Whether it be imitation or empathy, there is that same connection made between activity in the visually perceptive areas of the brain and activity in the motor areas in both cases.
DeleteGood reply, Hassanatou.
DeleteThis article covered some points that got me thinking that the intention-understanding-capacity we have as humans are probably the most important preamble that's needs to be tackled in the attempt to reverse engineer human capacities. Just as the premotor cortex that receive multiple cortical & peripheral inputs then outputs to direct behaviours, there might be something(e.g. intention to move my hand to reach for my tea) before the action of reaching the tea cup, there might be a pre-behaviour capacity that a T2 cannot obtain but a T3/T4 can.
ReplyDeleteWe have agreed upon that the mirror mechanism does not appear to be reducible to simply one-on-one connections, I couldn't help but think how would we tackle the question of understanding the reasoning/intention that's not observable. Maybe using EEG is not a bad idea since it's not invasive?
Monica, yes, the prior neural activity planning a movement pattern is similar to that of executing the movement pattern. But the unique thing about mirror capacity is the similarity or connection between the motor pattern and the sensory one of seeing it executed.
DeleteA T2 cannot move (or see) at all.
Everything an organism can DO is in principle observable, including brain activity (T4). What’s not observable is whether and what an organism feels. (But see Fodor about observing brain activity patterns.)
Mirror neurons (MN) seem to facilitate social learning of what we both do and feel, with special attention paid to whatever intention underlies these observed processes. Perhaps by reverse-engineering the mirror mechanism we can shed light on the connection between doing an action and what it feels like to do that action. Additionally, perhaps we can make a machine more "human" by reverse-engineering the mirror mechanism thanks to its empathetic function. Thinking about this concept reminded me of a quote by the anthropologist, Margaret Mead: "Helping someone else through difficulty is where civilisation starts." Mead is referring to evidence found of an ancient human femur that had broken and healed, an injury that's fatal in the animal kingdom and, therefore, reflective of a turning point in human evolution. Since monkeys also have MNs, I'm sure this turning point was not the advent of the human mirror mechanism, however, maybe it was the evolution of a certain sophistication of our MN circuity that enabled such empathy and communication to take place.
ReplyDeletePolly, what needs to be reverse-engineered is the capacity to do (cognitive things). Whether this needs to include T4 is another question.
DeleteThe conclusion of this article correctly observed that evaluating neural activation patterns and studying mirror neurons alone do not tell us how our empathy, or action and intention understanding functions. We can only hypothesize the mechanism of mirror neurons for now, and how they contribute to these facilitating these functions. Upon reflecting on the Fodor article after reading this one, I begin to rethink his ‘anti-empiricist’ stance of the relevance of neuroscientific research. I mentioned in my comment in 4(b) that “advancing brain-based research tells us what the brain is doing when we perform certain cognitive function, but fails to address why and how it does this”. But although there is no functional information about how/why mirror capacities work yet, this does not mean that there won’t be in the future. Consequently, this information would be incredibly valuable in terms of understanding how our neurons are able to perform these ‘mirroring capacities’, and has the potential to be ground-breaking.
ReplyDeleteSara, yes, there’s always the chance that the neural mechanism will be understood eventually. But MN correlations already suggest that mirror-capacities share similar mechanisms, which is useful to know for those who are trying to model mirror-capacities.
DeleteDiscovering the mechanism of mirror neurons and the link between motor actions with purpose and understanding was what stood out to me. What they found is that with a goal not only could action-understanding be provoked with motor actions but that there could be a combination/an even greater general understanding
ReplyDeleteI assume this would relate to the problem that comes with simulations. While the action can be mimicked, the “intentionality” that is missing here is the goal of such action. It broadened my focus a little on what it means to understand. So far, discussion has been that there is a barrier in understanding as it feels like something to understand or it feels like something to do things. I think this article provides another approach of not just looking to replicate but also in expectations of the resulting action and how that could result in that needed "feeling".
Hassanatou, read prior replies on the difference between language-understanding and action “understanding.” Both are felt, but meaning (semantics) is unique to language.
DeleteThe authors of this reading say that the main functions of mirror neurons might be an 'ill-posed question.' In fact, mirror neurons - which provides a way to map the sensory inputs to the motor system - might have no functions on its own at all but it contributes to a vast variety of functions. But back to the most basic question, what is exactly a function of a neural area? Is it equal to the human behaviour when it is active? I think the mirror neurons are merely 'enabling' humans to imitate other humans' behaviour instead of the imitation itself. As the encyclopedia says, its activeness 'is involved' in intention understanding but is not intention understanding itself. Each capacity of humans lies within complex pathways with interplays between brain areas and sensory organs and is on a higher, more macroscopic level than individual brain areas. So, does understanding mirror neurons mean understanding intention understanding, empathy, or language? I do not think so. Meanwhile, understanding basic structures is surely crucial for understanding higher-order mechanisms, but that's not all.
ReplyDeleteI agree with your claim that mere activation of a certain group of neurons is not equal to the human behaviour produced when that group is active. Or in other words, that the neural activity is not sufficient enough to explain how the behaviour is being done. Although I do find it hard to digest the following two excerpts from the reading simultaneously: mirror neurons "underlie a variety of functions" yet "do not have a unique functional role". Maybe I am just getting caught up in specific word use, because if the meaning of "a functional role" is something like, being wholly responsible for empathy, then I would of course disagree. But I think most people would take it to mean that it is sufficient enough for the meaning of "a functional role" to mean that mirror neurons play a part, along with other groups of neurons, in producing empathic behaviours.
DeleteHan, until a cognitive capacity has been successfully reverse-engineered we don’t understand its causal mechanism. Neural properties are always welcome, but what cogsci is seeking is a causal mechanism.
DeleteTyler, of course the brain causes DOing, Doing-capacity and FEELing. But knowing that a box causes X does not explain HOW the box causes X. For that you have to reverse-engineer how the box is doing it. (And that’s easier for DOing than for FEELing.
I understand that mirror neuron fire both when an animal act ad when the animal observes the same action performed by another. However, I'm wondering, will a mirror neuron fire when a monkey observes its own movement, or will a mirror neuron also fire when the monkey can't see its own movement? Do the findings of mirror neuron means that we internally replay the action we view in another before we act? Throughout the article, I'm not really convinced that there's any direct evidence that mirror neurons exist in the human brain. For example, it seems like the neuron will fire when actions are observed and mimicked(do the same action), but what about when they were mimicked first and then observed?
ReplyDeleteThis is a really interesting point, since one would think that if these neurons truly acted as “mirrors”, then a monkey should not have to be able to observe its own movement in order to imitate the movement of another human or monkey. Likely, they would be able to create a mental representation of the movement in their head, and then use this representation in order to recreate the motion using their own muscles. I’m a little confused by what you mean when you say if actions are “mimicked first and then observed”, since from my understanding, the whole point of mirror neurons is that they only allow the monkey to mimic the movement when it is observed, so how could the order be reversed?
DeleteMNs fire both when someone does an action of a certain shape or when someone else does an action of the same shape. Therefore they presumably fire when both are doing it at the same time too. Correlations are just correlations. It is the firing when the shape is motor or sensory that is the “mirroring.”
DeleteI thought this reading was interesting in providing a link between our neural systems and the concept of mirror neurons. It explains the functions and mechanisms under our capacity to imitate and replicate someone’s movements. This provides a ‘how’ in the physical property sense, however it doesn’t tackle the broader question related to the brain which is “how do we think?” Also, I found this extract particularly striking as it reminded me of the difference between computation or symbol manipulation and actual understanding of the objects manipulated. I thought the idea was similar to the Chinese Room Argument by Searle.
ReplyDelete“A mere visual perception, without involvement of the motor system would only provide a description of the visible aspects of the movements. It would not give, however, information on the intrinsic components of the observed action, on what it means to do the action, and on the links between the observed action and other actions related to it. This can be achieved only if the observed action is mapped onto the motor system of the observer.”
See reply above.
DeleteThis reading was quite enlightening about how MNs have been studied and hypothesized and analyzed in the past decades. Though the functional role of the mirror neuron system still is to be understood. I found it interesting that one of the highest correlated researchs to MNs was in conjunction of Autism. Probably due to autism both being a hot topic to speak about but also probably due to the social cognitive aspect of it. Many of my classmates have raised interesting points about MNs and I don't really have an interesting point of my own. I just have a personal interest in Psychopathology in relation to MNs to be particularly interesting and to be explored more may highlight some points about MNs and interneuron MNs.
ReplyDeleteMN activity correlation with autism is no more explanatory than its correlation with the sensory and motor shape of a movement. It doesn't explain how the brain produces (or loses) mirror-capacities.
DeleteThe most fascinating part of this article for me was definitely how mirror neurons have been implicated via fMRI experiments to be functionally involved in emotion and empathy, because it had never occurred to me before that neurons responsible for motor actions could also be directly involved in what is often considered higher level cortical function. Emotion, and in particular empathy, in practically all psychology and neuroscience courses I have taken, has always been talked about only in terms of the prefrontal lobe, with the anterior insula and anterior cingulate never even being mentioned. The idea that the insula is active both when experiencing something and when observing the reaction to that same thing also emphasizes strong connection that the motor regions of the brain have with regions involved in complicated cognitive functions.
ReplyDeleteYes, MNs reveal the perception/production mirroring by the same neurons. But there is no explanation of HOW the neurons do it.
Delete
ReplyDeleteI feel like answering this question really comes back to a singular idea which is that there is no physiological fingerprint of emotions, particularly, those arising in social contexts and that in the lab, this is difficult to replicate. In a lot of the discussions of cognition, even in class, we intuitively feel as though cognition should relate to emotion even though that may not necessarily be the case. However, with that intuition, it then becomes difficult to accept that neuroscience can explain cognition if it cannot provide a physiological framework or "fingerprint" to describe something we feel is so central to cognition (even if in reality, it isn't)
MNs don't explain how we DO mirror tasks like imitating an action we see by another person; so they don't explain how we FEEL the same emotion as another person we see emoting either.
DeleteIn relation to this question, I feel as though a lot of discordances in cognitive versus neuroscientific explanations arise from the fact that the fields and their effects are regarded as divergent in some senses. For instance, the (Bonini) article writes "one of the central remaining controversies concerns the ontogenetic origin of MNs, that is, whether they are innate or forged by learning processes. " which asks a very clear-cut question which likely has an answer that isn't very clearcut. More generally, cognition as we now know it is still a very broadly defined term and one that we continue to debate. Some people believe that a central component to cognition is emotion, some don't. In this way, it becomes difficult to find objective data in support of questions relating to cognition whereas neuroscience relies on this objectivity in making claims. In other words, a lot of the controversy arises from the uncertainty in defining and grounding cognition when held against the need for objectivity in neuroscience.
ReplyDeleteMN activity is correlated with both doing and seeing an action of the same "shape." There is no reason it cannot be tested whether the correlation is inborn or learned.
DeletePart of this reading dealt with the idea that human speech evolved from a gestural communication system involving mirror neurons: we began with gestures grounded to some degree in real-world referents where mirror neurons in the other party would then facilitate understanding by activating the same motor pathways as a gesture or interaction with an object. While I'm not immediately fully convinced by this hypothesis for the origin of language, we do know that to process speech we make significant use of non-verbal information as well. As an example, the McGurk effect (https://www.youtube.com/watch?v=2k8fHR9jKVM) shows how listening to audio of someone saying a sound but while seeing a video of someone produce another sound tends to influence us to perceive the second sound. It would be interesting to find whether this phenomenon is connected to mirror neuron activation, i.e. whether mirror neurons facilitate the firing of the same motor pathways used in producing the sounds we are seeing and whether this thereby influences our perceptions.
ReplyDeleteYour question about the McGurk effect is interesting, and maybe testable, It would be interesting to test phoneme recognition in deaf people: Are the correlations just visual-motor or visual-auditory-motor? The (hypothetical) relation to gesture and the evolution of language is more speculative, but could perhaps be tested in signing by deaf people who were born deaf or became deaf (and learned sign language) later in life.
DeleteA Scholar search shows that several people have in fact studied the McGurk effect in relation to mirror neurons. While I haven't found any of the specific sort you suggested, I was intrigued by a few studies that test whether an equivalent of the McGurk effect can be found when individuals use the Tadoma method, used by some Deaf people to perceive spoken language, wherein the receiver puts their fingers on the lips, jaw, and throat of the speaker to perceive their speech. For the moment results appear inconclusive, but it seems to me an interesting line of thinking.
DeleteI had a difficult time understanding the relation between mirror neurons and language, but I think the concluding paragraph and other comments has clarified my understanding. In the last paragraph of the reading, the authors write that “the mirror mechanism does not explain by itself the enormous complexity of speech, yet it solves one of the fundamental difficulties for understanding language evolution.” I think this is exactly it– mirror neurons are fundamental in theorizing about language evolution. Pantomime is NOT language, but in the evolution of speech these gestures were crucial in communicating a message from the sender to the receiver without the use of arbitrary symbols. In language, these arbitrary symbols are what we know as words/labels for everything that we see around us. We could argue that when languages were coming to be, the words used to label things were ‘arbitrary,’ but the gestures associated with these words were less-so. They helped make what is “valid for the sender of a message valid for the receiver.”
ReplyDeleteMirror neurons are very interesting in that we don't quite know their specific use, but researchers have found various things that they seem to be involved in. Since they discharge selectively during the observation of motor acts when they're embedded in a given action (i.e. grasping an apple to eat it), I assume that mirror neurons play a role in the inferences we make in our daily lives to understand the intention of others' actions. Knowing what it feels like to be in a particular state that another person is in gives us the information about what exactly that person feels and potentially what their next move may be. It's interesting to learn that autistic patients have an impaired mirror neuron system because one could guess that such an impairment would significantly affect learning ability. If you can't infer what other people are going to do based on their behaviour, it will be difficult to understand the world around you.
ReplyDeleteI found it fascinating how the “meaning” or perhaps “intention” of the act plays such an important role on these neurons. Furthermore, this article has made me realize how all of the functions from the mirror neurons are interconnected. That is, empathy, something that is essential when establishing human connections, can be linked with things such as autism which carries the lack of understanding the actions of others. Lastly, this article has also shown me a different way for thinking about language evolution, which I had previously thought had evolved from animals such as monkeys making noise. To conclude, I would like to say that I really enjoyed this reading and that it was very informative and easy to understand.
ReplyDeleteThis reading on mirror neurons was really interesting. From my understanding mirror neurons enable individuals to execute their mirror capacities. For example, when infants observe the actions of people in their environment and copy their behaviour, is an instance where mirror neurons are discharged.
ReplyDeleteIn the case of autism, the reading touches on how for autistic children they are probably unable to relate to other people or life situation due to their mirror neuron system not functioning normally. I used the infant example earlier, so would autistic infant would likely not copy the actions of the people in their environment because of damage of the mirror system which is needed to understand other people.
This article is an interesting look at the role of mirror neurons in action understanding, imitation, intention understanding, empathy, and language understanding/perception. This reading is particularly focused on how mirror neurons are activated through visual observation. It prompted me to consider how mirror neurons seem to be able to be activated through means other than visual observation. For example, people are able to empathize with characters that they read in novels, which suggests that mirror neurons (in this case, the mirror capacity of empathy) may be activated through language alone. Similarly, when kids learn to speak they imitate speech (non motor system imitation) by attempting to repeat the sounds they hear another person makes, though it's unclear how much visually observing someone making a sound influences ability to imitate that sound.
ReplyDelete